2010-02-04 21:30:06

by Josh Holland

[permalink] [raw]
Subject: [PATCH] Staging: rar: fixed up rar_driver.{h,c}

This is a patch to the rar_driver.c and rar_driver.h files to remove
style issues found by the checkpatch.pl script.

Signed-off-by: Josh Holland <[email protected]>
---
drivers/staging/rar/rar_driver.c | 92 ++++++++++++++++----------------------
drivers/staging/rar/rar_driver.h | 72 +++++++++++------------------
2 files changed, 67 insertions(+), 97 deletions(-)

diff --git a/drivers/staging/rar/rar_driver.c b/drivers/staging/rar/rar_driver.c
index d85d189..886a819 100644
--- a/drivers/staging/rar/rar_driver.c
+++ b/drivers/staging/rar/rar_driver.c
@@ -68,7 +68,8 @@ static void __exit rar_exit_handler(void);
/*
function that is activated on the successfull probe of the RAR device
*/
-static int __devinit rar_probe(struct pci_dev *pdev, const struct pci_device_id *ent);
+static int __devinit rar_probe(struct pci_dev *pdev,
+ const struct pci_device_id *ent);

static struct pci_device_id rar_pci_id_tbl[] = {
{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x4110) },
@@ -86,9 +87,7 @@ static struct pci_driver rar_pci_driver = {

/* This function is used to retrieved RAR info using the IPC message
bus interface */
-static int memrar_get_rar_addr(struct pci_dev* pdev,
- int offset,
- u32 *addr)
+static int memrar_get_rar_addr(struct pci_dev *pdev, int offset, u32 *addr)
{
/*
* ======== The Lincroft Message Bus Interface ========
@@ -140,23 +139,19 @@ static int memrar_get_rar_addr(struct pci_dev* pdev,
| (offset << 8)
| (LNC_MESSAGE_BYTE_WRITE_ENABLES << 4);

- printk(KERN_WARNING "rar- offset to LNC MSG is %x\n",offset);
+ printk(KERN_WARNING "rar- offset to LNC MSG is %x\n", offset);

if (addr == 0)
return -EINVAL;

/* Send the control message */
- result = pci_write_config_dword(pdev,
- LNC_MCR_OFFSET,
- message);
+ result = pci_write_config_dword(pdev, LNC_MCR_OFFSET, message);

- printk(KERN_WARNING "rar- result from send ctl register is %x\n"
- ,result);
+ printk(KERN_WARNING "rar- result from send ctl register is %x\n",
+ result);

if (!result)
- result = pci_read_config_dword(pdev,
- LNC_MDR_OFFSET,
- addr);
+ result = pci_read_config_dword(pdev, LNC_MDR_OFFSET, addr);

printk(KERN_WARNING "rar- result from read data register is %x\n",
result);
@@ -170,9 +165,7 @@ static int memrar_get_rar_addr(struct pci_dev* pdev,
return 0;
}

-static int memrar_set_rar_addr(struct pci_dev* pdev,
- int offset,
- u32 addr)
+static int memrar_set_rar_addr(struct pci_dev *pdev, int offset, u32 addr)
{
/*
* ======== The Lincroft Message Bus Interface ========
@@ -225,23 +218,19 @@ static int memrar_set_rar_addr(struct pci_dev* pdev,
| (offset << 8)
| (LNC_MESSAGE_BYTE_WRITE_ENABLES << 4);

- printk(KERN_WARNING "rar- offset to LNC MSG is %x\n",offset);
+ printk(KERN_WARNING "rar- offset to LNC MSG is %x\n", offset);

if (addr == 0)
return -EINVAL;

/* Send the control message */
- result = pci_write_config_dword(pdev,
- LNC_MDR_OFFSET,
- addr);
+ result = pci_write_config_dword(pdev, LNC_MDR_OFFSET, addr);

- printk(KERN_WARNING "rar- result from send ctl register is %x\n"
- ,result);
+ printk(KERN_WARNING "rar- result from send ctl register is %x\n",
+ result);

if (!result)
- result = pci_write_config_dword(pdev,
- LNC_MCR_OFFSET,
- message);
+ result = pci_write_config_dword(pdev, LNC_MCR_OFFSET, message);

printk(KERN_WARNING "rar- result from write data register is %x\n",
result);
@@ -284,35 +273,31 @@ static int memrar_init_rar_params(struct pci_dev *pdev)
/* struct pci_dev *pdev = pci_get_bus_and_slot(0, PCI_DEVFN(0,0)); */

if (pdev == NULL)
- return -ENODEV;
+ return -ENODEV;

for (i = offsets; i != end; ++i, ++n) {
- if (memrar_get_rar_addr (pdev,
- (*i).low,
- &(rar_addr[n].low)) != 0
- || memrar_get_rar_addr (pdev,
- (*i).high,
- &(rar_addr[n].high)) != 0) {
- result = -1;
- break;
- }
+ if (memrar_get_rar_addr(pdev, (*i).low, &(rar_addr[n].low))
+ || memrar_get_rar_addr(pdev, (*i).high,
+ &(rar_addr[n].high))) {
+ result = -1;
+ break;
+ }
}

/* Done accessing the device. */
/* pci_dev_put(pdev); */

if (result == 0) {
- if(1) {
- size_t z;
- for (z = 0; z != MRST_NUM_RAR; ++z) {
- printk(KERN_WARNING "rar - BRAR[%Zd] physical address low\n"
- "\tlow: 0x%08x\n"
- "\thigh: 0x%08x\n",
- z,
- rar_addr[z].low,
- rar_addr[z].high);
- }
- }
+ size_t z;
+ for (z = 0; z != MRST_NUM_RAR; ++z) {
+ printk(KERN_WARNING "rar - "
+ "BRAR[%Zd] physical address low\n"
+ "\tlow: 0x%08x\n"
+ "\thigh: 0x%08x\n",
+ z,
+ rar_addr[z].low,
+ rar_addr[z].high);
+ }
}

return result;
@@ -321,7 +306,8 @@ static int memrar_init_rar_params(struct pci_dev *pdev)
/*
function that is activated on the successfull probe of the RAR device
*/
-static int __devinit rar_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
+static int __devinit rar_probe(struct pci_dev *pdev,
+ const struct pci_device_id *ent)
{
/* error */
int error;
@@ -347,7 +333,7 @@ static int __devinit rar_probe(struct pci_dev *pdev, const struct pci_device_id

/* Initialize the RAR parameters, which have to be retrieved */
/* via the message bus service */
- error=memrar_init_rar_params(rar_dev);
+ error = memrar_init_rar_params(rar_dev);

if (error) {
DEBUG_PRINT_0(RAR_DEBUG_LEVEL_EXTENDED,
@@ -396,10 +382,10 @@ MODULE_LICENSE("GPL");
* The function returns a 0 upon success or a -1 if there is no RAR
* facility on this system.
*/
-int get_rar_address(int rar_index,struct RAR_address_struct *addresses)
+int get_rar_address(int rar_index, struct RAR_address_struct *addresses)
{
if (registered && (rar_index < 3) && (rar_index >= 0)) {
- *addresses=rar_addr[rar_index];
+ *addresses = rar_addr[rar_index];
/* strip off lock bit information */
addresses->low = addresses->low & 0xfffffff0;
addresses->high = addresses->high & 0xfffffff0;
@@ -410,9 +396,9 @@ int get_rar_address(int rar_index,struct RAR_address_struct *addresses)
return -ENODEV;
}
}
+EXPORT_SYMBOL(get_rar_address);


-EXPORT_SYMBOL(get_rar_address);

/* The lock_rar function is ued by other device drivers to lock an RAR.
* once an RAR is locked, it stays locked until the next system reboot.
@@ -431,10 +417,10 @@ int lock_rar(int rar_index)
int result;
if (registered && (rar_index < 3) && (rar_index >= 0)) {
/* first make sure that lock bits are clear (this does lock) */
- working_addr=rar_addr[rar_index].low & 0xfffffff0;
+ working_addr = rar_addr[rar_index].low & 0xfffffff0;

/* now send that value to the register using the IPC */
- result=memrar_set_rar_addr(rar_dev,rar_index,working_addr);
+ result = memrar_set_rar_addr(rar_dev, rar_index, working_addr);
return result;
}

diff --git a/drivers/staging/rar/rar_driver.h b/drivers/staging/rar/rar_driver.h
index 3690f98..567cc7e 100644
--- a/drivers/staging/rar/rar_driver.h
+++ b/drivers/staging/rar/rar_driver.h
@@ -1,7 +1,7 @@
/* === RAR Physical Addresses === */
struct RAR_address_struct {
- u32 low;
- u32 high;
+ u32 low;
+ u32 high;
};

/* The get_rar_address function is used by other device drivers
@@ -19,7 +19,7 @@ struct RAR_address_struct {
* The function returns a 0 upon success or a -1 if there is no RAR
* facility on this system.
*/
-int get_rar_address(int rar_index,struct RAR_address_struct *addresses);
+int get_rar_address(int rar_index, struct RAR_address_struct *addresses);


/* The lock_rar function is ued by other device drivers to lock an RAR.
@@ -48,52 +48,36 @@ int lock_rar(int rar_index);
/* FUNCTIONAL MACROS */

/* debug macro without paramaters */
-#define DEBUG_PRINT_0(DEBUG_LEVEL , info) \
-do \
-{ \
- if(DEBUG_LEVEL) \
- { \
- printk(KERN_WARNING info); \
- } \
-}while(0)
+#define DEBUG_PRINT_0(DEBUG_LEVEL, info) \
+do { \
+ if (DEBUG_LEVEL) \
+ printk(KERN_WARNING info); \
+} while (0)

/* debug macro with 1 paramater */
-#define DEBUG_PRINT_1(DEBUG_LEVEL , info , param1) \
-do \
-{ \
- if(DEBUG_LEVEL) \
- { \
- printk(KERN_WARNING info , param1); \
- } \
-}while(0)
+#define DEBUG_PRINT_1(DEBUG_LEVEL, info , param1) \
+do { \
+ if (DEBUG_LEVEL) \
+ printk(KERN_WARNING info, param1); \
+} while (0)

/* debug macro with 2 paramaters */
-#define DEBUG_PRINT_2(DEBUG_LEVEL , info , param1, param2) \
-do \
-{ \
- if(DEBUG_LEVEL) \
- { \
- printk(KERN_WARNING info , param1, param2); \
- } \
-}while(0)
+#define DEBUG_PRINT_2(DEBUG_LEVEL, info, param1, param2) \
+do { \
+ if (DEBUG_LEVEL) \
+ printk(KERN_WARNING info, param1, param2); \
+} while (0)

/* debug macro with 3 paramaters */
-#define DEBUG_PRINT_3(DEBUG_LEVEL , info , param1, param2 , param3) \
-do \
-{ \
- if(DEBUG_LEVEL) \
- { \
- printk(KERN_WARNING info , param1, param2 , param3); \
- } \
-}while(0)
+#define DEBUG_PRINT_3(DEBUG_LEVEL, info, param1, param2, param3) \
+do { \
+ if (DEBUG_LEVEL) \
+ printk(KERN_WARNING info, param1, param2, param3); \
+} while (0)

/* debug macro with 4 paramaters */
-#define DEBUG_PRINT_4(DEBUG_LEVEL , info , param1, param2 , param3 , param4) \
-do \
-{ \
- if(DEBUG_LEVEL) \
- { \
- printk(KERN_WARNING info , param1, param2 , param3 , param4); \
- } \
-}while(0)
-
+#define DEBUG_PRINT_4(DEBUG_LEVEL, info, param1, param2, param3, param4) \
+do { \
+ if (DEBUG_LEVEL) \
+ printk(KERN_WARNING info, param1, param2, param3, param4); \
+} while (0)
--
1.6.3.3


2010-02-05 00:32:57

by Krzysztof Halasa

[permalink] [raw]
Subject: Re: [PATCH] Staging: rar: fixed up rar_driver.{h,c}

Josh Holland <[email protected]> writes:

> This is a patch to the rar_driver.c and rar_driver.h files to remove
> style issues found by the checkpatch.pl script.
>
> +++ b/drivers/staging/rar/rar_driver.c
> @@ -68,7 +68,8 @@ static void __exit rar_exit_handler(void);
> /*
> function that is activated on the successfull probe of the RAR device
> */
> -static int __devinit rar_probe(struct pci_dev *pdev, const struct pci_device_id *ent);
> +static int __devinit rar_probe(struct pci_dev *pdev,
> + const struct pci_device_id *ent);

It's agreed such changes make it worse. The 80-column "ERROR" should be
ignored, and it will be removed from checkpatch.

> - printk(KERN_WARNING "rar- result from send ctl register is %x\n"
> - ,result);
> + printk(KERN_WARNING "rar- result from send ctl register is %x\n",
> + result);

Also, here (and then) - I'd just make it a single line if you're changing
it. I'd be far from "unwrapping" all code across the kernel, though
(without otherwise changing the lines in question).

> + if (memrar_get_rar_addr(pdev, (*i).low, &(rar_addr[n].low))
> + || memrar_get_rar_addr(pdev, (*i).high,
> + &(rar_addr[n].high))) {
> + result = -1;
> + break;
> + }

Isn't the following a bit more readable?

+ if (memrar_get_rar_addr(pdev, i->low, &rar_addr[n].low) ||
+ memrar_get_rar_addr(pdev, i->high, &rar_addr[n].high)) {
+ result = -1;
+ break;
+ }

It doesn't make sense to split the printk, at least every single output
line printed shouldn't be broken into pieces (but perhaps one single
line for the whole printk() is best).
Also I like the post-increments (z++) more, but maybe it's just me.

> + size_t z;
> + for (z = 0; z != MRST_NUM_RAR; ++z) {
> + printk(KERN_WARNING "rar - "
> + "BRAR[%Zd] physical address low\n"
> + "\tlow: 0x%08x\n"
> + "\thigh: 0x%08x\n",
> + z,
> + rar_addr[z].low,
> + rar_addr[z].high);


> -#define DEBUG_PRINT_0(DEBUG_LEVEL , info) \
> -do \
> -{ \
> - if(DEBUG_LEVEL) \
> - { \
> - printk(KERN_WARNING info); \
> - } \
> -}while(0)
> +#define DEBUG_PRINT_0(DEBUG_LEVEL, info) \
> +do { \
> + if (DEBUG_LEVEL) \
> + printk(KERN_WARNING info); \
> +} while (0)

Also I think moving these backslashes to the right of the macro code is
preferred, isn't it?

Just my 0.01$CURRENCY as usual :-)
--
Krzysztof Halasa

2010-02-05 17:23:08

by Greg KH

[permalink] [raw]
Subject: Re: [PATCH] Staging: rar: fixed up rar_driver.{h,c}

On Fri, Feb 05, 2010 at 01:32:52AM +0100, Krzysztof Halasa wrote:
> Josh Holland <[email protected]> writes:
>
> > This is a patch to the rar_driver.c and rar_driver.h files to remove
> > style issues found by the checkpatch.pl script.
> >
> > +++ b/drivers/staging/rar/rar_driver.c
> > @@ -68,7 +68,8 @@ static void __exit rar_exit_handler(void);
> > /*
> > function that is activated on the successfull probe of the RAR device
> > */
> > -static int __devinit rar_probe(struct pci_dev *pdev, const struct pci_device_id *ent);
> > +static int __devinit rar_probe(struct pci_dev *pdev,
> > + const struct pci_device_id *ent);
>
> It's agreed such changes make it worse. The 80-column "ERROR" should be
> ignored, and it will be removed from checkpatch.
>
> > - printk(KERN_WARNING "rar- result from send ctl register is %x\n"
> > - ,result);
> > + printk(KERN_WARNING "rar- result from send ctl register is %x\n",
> > + result);
>
> Also, here (and then) - I'd just make it a single line if you're changing
> it. I'd be far from "unwrapping" all code across the kernel, though
> (without otherwise changing the lines in question).

No, this is fine, no problem with this change.

> > + if (memrar_get_rar_addr(pdev, (*i).low, &(rar_addr[n].low))
> > + || memrar_get_rar_addr(pdev, (*i).high,
> > + &(rar_addr[n].high))) {
> > + result = -1;
> > + break;
> > + }
>
> Isn't the following a bit more readable?
>
> + if (memrar_get_rar_addr(pdev, i->low, &rar_addr[n].low) ||
> + memrar_get_rar_addr(pdev, i->high, &rar_addr[n].high)) {
> + result = -1;
> + break;
> + }

The latter is nicer, but it doesn't really matter :)

> It doesn't make sense to split the printk, at least every single output
> line printed shouldn't be broken into pieces (but perhaps one single
> line for the whole printk() is best).
> Also I like the post-increments (z++) more, but maybe it's just me.
>
> > + size_t z;
> > + for (z = 0; z != MRST_NUM_RAR; ++z) {
> > + printk(KERN_WARNING "rar - "
> > + "BRAR[%Zd] physical address low\n"
> > + "\tlow: 0x%08x\n"
> > + "\thigh: 0x%08x\n",
> > + z,
> > + rar_addr[z].low,
> > + rar_addr[z].high);
>
>
> > -#define DEBUG_PRINT_0(DEBUG_LEVEL , info) \
> > -do \
> > -{ \
> > - if(DEBUG_LEVEL) \
> > - { \
> > - printk(KERN_WARNING info); \
> > - } \
> > -}while(0)
> > +#define DEBUG_PRINT_0(DEBUG_LEVEL, info) \
> > +do { \
> > + if (DEBUG_LEVEL) \
> > + printk(KERN_WARNING info); \
> > +} while (0)
>
> Also I think moving these backslashes to the right of the macro code is
> preferred, isn't it?


It doesn't matter all that much.

Overall this looks fine, I'll queue it up.

thanks,

greg k-h

2010-02-05 21:32:55

by Krzysztof Halasa

[permalink] [raw]
Subject: Re: [PATCH] Staging: rar: fixed up rar_driver.{h,c}

Greg KH <[email protected]> writes:

> It doesn't matter all that much.
>
> Overall this looks fine, I'll queue it up.

It's an improvement, sure. Doing improvements in smaller, incomplete
steps is right, sure. But it can still be improved further.
--
Krzysztof Halasa

2010-02-05 21:58:44

by Josh Holland

[permalink] [raw]
Subject: [PATCH 2/2] Staging: rar: More style changes to rar_driver.c

Following advice from Krzysztof Halasa, I made a few changes to the
style bits in rar_driver.c

Signed-off-by: Josh Holland <[email protected]>
---
drivers/staging/rar/rar_driver.c | 21 +++++++++------------
1 files changed, 9 insertions(+), 12 deletions(-)

diff --git a/drivers/staging/rar/rar_driver.c b/drivers/staging/rar/rar_driver.c
index 886a819..a905192 100644
--- a/drivers/staging/rar/rar_driver.c
+++ b/drivers/staging/rar/rar_driver.c
@@ -68,8 +68,7 @@ static void __exit rar_exit_handler(void);
/*
function that is activated on the successfull probe of the RAR device
*/
-static int __devinit rar_probe(struct pci_dev *pdev,
- const struct pci_device_id *ent);
+static int __devinit rar_probe(struct pci_dev *pdev, const struct pci_device_id *ent);

static struct pci_device_id rar_pci_id_tbl[] = {
{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x4110) },
@@ -102,7 +101,7 @@ static int memrar_get_rar_addr(struct pci_dev *pdev, int offset, u32 *addr)
* 2. [23:16]: Port
* 3. [15:8]: Register Offset
* 4. [7:4]: Byte Enables (use 0xF to set all of these bits
- * to 1)
+ * to 1)
* 5. [3:0]: reserved
*
* Read (0xD0) and write (0xE0) opcodes are written to the
@@ -147,8 +146,7 @@ static int memrar_get_rar_addr(struct pci_dev *pdev, int offset, u32 *addr)
/* Send the control message */
result = pci_write_config_dword(pdev, LNC_MCR_OFFSET, message);

- printk(KERN_WARNING "rar- result from send ctl register is %x\n",
- result);
+ printk(KERN_WARNING "rar- result from send ctl register is %x\n", result);

if (!result)
result = pci_read_config_dword(pdev, LNC_MDR_OFFSET, addr);
@@ -180,7 +178,7 @@ static int memrar_set_rar_addr(struct pci_dev *pdev, int offset, u32 addr)
* 2. [23:16]: Port
* 3. [15:8]: Register Offset
* 4. [7:4]: Byte Enables (use 0xF to set all of these bits
- * to 1)
+ * to 1)
* 5. [3:0]: reserved
*
* Read (0xD0) and write (0xE0) opcodes are written to the
@@ -276,9 +274,8 @@ static int memrar_init_rar_params(struct pci_dev *pdev)
return -ENODEV;

for (i = offsets; i != end; ++i, ++n) {
- if (memrar_get_rar_addr(pdev, (*i).low, &(rar_addr[n].low))
- || memrar_get_rar_addr(pdev, (*i).high,
- &(rar_addr[n].high))) {
+ if (memrar_get_rar_addr(pdev, (*i).low, &(rar_addr[n].low)) ||
+ memrar_get_rar_addr(pdev, (*i).high, &(rar_addr[n].high))) {
result = -1;
break;
}
@@ -289,9 +286,9 @@ static int memrar_init_rar_params(struct pci_dev *pdev)

if (result == 0) {
size_t z;
- for (z = 0; z != MRST_NUM_RAR; ++z) {
- printk(KERN_WARNING "rar - "
- "BRAR[%Zd] physical address low\n"
+ for (z = 0; z != MRST_NUM_RAR; z++) {
+ printk(KERN_WARNING
+ "rar - BRAR[%Zd] physical address low\n"
"\tlow: 0x%08x\n"
"\thigh: 0x%08x\n",
z,
--
1.6.3.3

2010-02-17 23:18:15

by Greg KH

[permalink] [raw]
Subject: Re: [PATCH 2/2] Staging: rar: More style changes to rar_driver.c

On Fri, Feb 05, 2010 at 09:58:26PM +0000, Josh Holland wrote:
> Following advice from Krzysztof Halasa, I made a few changes to the
> style bits in rar_driver.c

Can you respin both of these patches in 2 days, against the linux-next
tree, as the rar driver author has sent in a bunch of "rename these
files" type patches, which caused these to not apply anymore.

thanks,

greg k-h

2010-02-20 11:45:24

by Josh Holland

[permalink] [raw]
Subject: Re: [PATCH 2/2] Staging: rar: More style changes to rar_driver.c

On Wed, Feb 17, 2010 at 02:40:46PM -0800, Greg KH wrote:
> On Fri, Feb 05, 2010 at 09:58:26PM +0000, Josh Holland wrote:
> > Following advice from Krzysztof Halasa, I made a few changes to the
> > style bits in rar_driver.c
>
> Can you respin both of these patches in 2 days, against the linux-next
> tree, as the rar driver author has sent in a bunch of "rename these
> files" type patches, which caused these to not apply anymore.
How do I do that? I can't see a linux-next branch in my local repo, and
I'd rather not download a whole new one...

$ git remote show origin
* remote origin
URL:
git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git
HEAD branch: master
Remote branch:
master tracked
Local branch configured for 'git pull':
master merges with remote master
Local ref configured for 'git push':
master pushes to master (up to date)


--
Josh "dutchie" Holland <[email protected]>
http://www.joshh.co.uk/
http://twitter.com/jshholland
http://identi.ca/jshholland

2010-02-24 03:40:05

by Greg KH

[permalink] [raw]
Subject: Re: [PATCH 2/2] Staging: rar: More style changes to rar_driver.c

On Sat, Feb 20, 2010 at 11:36:07AM +0000, Josh Holland wrote:
> On Wed, Feb 17, 2010 at 02:40:46PM -0800, Greg KH wrote:
> > On Fri, Feb 05, 2010 at 09:58:26PM +0000, Josh Holland wrote:
> > > Following advice from Krzysztof Halasa, I made a few changes to the
> > > style bits in rar_driver.c
> >
> > Can you respin both of these patches in 2 days, against the linux-next
> > tree, as the rar driver author has sent in a bunch of "rename these
> > files" type patches, which caused these to not apply anymore.
> How do I do that? I can't see a linux-next branch in my local repo, and
> I'd rather not download a whole new one...
>
> $ git remote show origin
> * remote origin
> URL:
> git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git
> HEAD branch: master
> Remote branch:
> master tracked
> Local branch configured for 'git pull':
> master merges with remote master
> Local ref configured for 'git push':
> master pushes to master (up to date)

Google for "linux-next FAQ" for how to set this up properly.

thanks,

greg k-h